Bug #1712

pulseaudio fails to build due to mlockall missed in libc

Added by dennis.melentyev about 3 years ago. Updated about 3 years ago.

Status:Closed Start date:
Priority:Normal Due date:
Assignee:- % Done:

0%
Category:-
Target version:-

Description

Hi,

I tried to build pulseaudio from pkgsrc and it failed because of
mlockall() is not in libc.
As far as I understand, that part is not critical and I modified
patch-ab as attached.

Another thing is: we do have mlockall syscall in mman.h AND some code
in kernel sources about this.
Shouldn't we put relevant bits into libc? Sorry, I'm not going to address this.

The system:
dennis@dfly (xterm-color) > gcc -v
Using built-in specs.
Target: i386-pc-dragonflybsd
Configured with: DragonFly/i386 system compiler
Thread model: posix
gcc version 4.1.2 (DragonFly)

dennis@dfly (xterm-color) > uname -a
DragonFly dfly.melent.local 2.5.1-DEVELOPMENT DragonFly
v2.5.1.1031.g8ded21-DEVELOPMENT #1: Wed Mar 24 01:41:57 EET 2010
root@dfly.melent.local:/usr/obj/usr/src/sys/GENERIC i386

patch-ab (974 Bytes) dennis.melentyev, 03/31/2010 11:17 pm

Related todos

History

#1
Updated by justin about 3 years ago

On Wed, March 31, 2010 7:15 pm, Dennis Melentyev wrote:
> Hi,
>
> I tried to build pulseaudio from pkgsrc and it failed because of
> mlockall() is not in libc.
> As far as I understand, that part is not critical and I modified
> patch-ab as attached.

Have you sent this in for pkgsrc via NetBSD's bug reporting page? I had
very quick results yesterday asking for sun-jdk6/sun-jre6 to be unmasked
for DragonFly.

#2
Updated by ahuete.devel about 3 years ago

Hi,

Any update on this? I've just tested the build today again and it
stills fails so I guess the patch hasn't been submitted in a PR.

Cheers,
Antonio Huete

2010/4/1 Justin C. Sherrill <justin@shiningsilence.com>:
> On Wed, March 31, 2010 7:15 pm, Dennis Melentyev wrote:
>> Hi,
>>
>> I tried to build pulseaudio from pkgsrc and it failed because of
>> mlockall() is not in libc.
>> As far as I understand, that part is not critical and I modified
>> patch-ab as attached.
>
> Have you sent this in for pkgsrc via NetBSD's bug reporting page?  I had
> very quick results yesterday asking for sun-jdk6/sun-jre6 to be unmasked
> for DragonFly.
>
>
>

#3
Updated by vsrinivas about 3 years ago

So mlockall/munlockall is a no-op in the kernel, but the syscall handler is
there. We should probably still expose it from libc and have the kernel return
ENOSYS, right?

#4
Updated by ahuete.devel about 3 years ago

Hi,

That sounds like a good idea to me, +1.

Cheers,
Antonio Huete

On 04/11/10 18:00, Venkatesh Srinivas (via DragonFly issue tracker) wrote:
> Venkatesh Srinivas <me@acm.jhu.edu> added the comment:
>
> So mlockall/munlockall is a no-op in the kernel, but the syscall handler is
> there. We should probably still expose it from libc and have the kernel return
> ENOSYS, right?
>
> _____________________________________________________
> DragonFly issue tracker <bugs@lists.dragonflybsd.org>
> <http://bugs.dragonflybsd.org/issue1712&gt;
> _____________________________________________________
>
>

#5
Updated by vsrinivas about 3 years ago

Patch to expose mlockall/munlockall, have them return ENOSYS.

http://endeavour.zapto.org/dfly/mlockall.diff

#6
Updated by tuxillo about 3 years ago

Hi,

Committed in 5b6d478465a99af56589ef42d9f7a5d90d9adce4

This should fix the build for audio/pulseaudio.

Thanks Venkatesh!

Regards,
Antonio Huete

#7
Updated by vsrinivas about 3 years ago

Perhaps its time to close this bug report?

Also available in: Atom PDF